[论文] Strategist: Evolving Reasoning with Agents

论文概要

研究领域: NLP 作者: Taiming Lu, Qi Xu, Zhigang Hua, Zhuang Liu 发布时间: 2026-11-02 arXiv: 2611.03618

中文摘要

模型只是 AI 智能体系统的一半，另一半是推理策略：将模型调用转化为响应的推理过程。今天这另一半靠手写，且在一个策略上积累的经验从不会迁移到下一个。我们提出 Strategist——一个跨任务进化推理策略的元智能体。每个策略被写成一份「蓝图」（blueprint）——可执行代码，可被其他策略复用，所有蓝图构成一个库。给定任务与一个小规模开发集，Strategist 检索相关蓝图、编写候选策略、在真实样本上测试、并根据结果修订。每个候选都会进入库中，使后续任务能建立在更早的发现之上。在 30 个基准上，Strategist 在 24 个上击败最强基线，平均准确率与基线持平（74.7% vs 74.0%），但推理成本低 39%。其策略还能无需进一步搜索就迁移到未见任务与其他模型。AI 系统中过去从未自我改进的那一半，现在开始改进了。

原文摘要

A model is only half of an AI agent system. The other half is the inference strategy: the reasoning procedure that turns model calls into a response. Today that second half is written by hand, and the experience of building one strategy never carries to the next. We introduce Strategist, a meta-agent that evolves inference strategies across tasks. Each strategy is written as a blueprint,executable code that other strategies can reuse, and together the blueprints form a library. Given a task and a small development set, Strategist retrieves related blueprints, writes candidate strategies, tests them on real samples, and revises them from the results. Every candidate then enters the library, so later tasks build on earlier discoveries. On 30 benchmarks, Strategist beats the strongest baselin...

#论文 #arXiv #NLP #小凯